Wrong field names when dot in names

[PhunkyBob]

I would like to retrieve all elements of a PDF form and display for each element it's content.
When I read fields having a "." in its name, the "key" returned is only the last part of the key, and the value is the value of the last element having this (wrong) key.

Environment

Windows 11 (Windows-10-10.0.22623-SP0)
Python 3.10
pypdf 3.2.0 / PyPDF2-2.11.2

Code + PDF

I have a PDF file with the following fields:

• customer.name
• customer.lastname
• other_field
• company.name

fields_with_dots.pdf

I want to get all fields and the corresponding values.

from PyPDF2 import PdfFileReader

if __name__ == "__main__":
    pdf_file_name = "fields_with_dots.pdf"
    with open(pdf_file_name, "rb") as pdfobject:
        pdf = PdfFileReader(pdfobject)
        fields = pdf.getFormTextFields()
        for k, v in fields.items():
            print(f"{k:32} : {v}")

What I expected

customer.name               : My Name
customer.lastname         : My Lastname
other_field                      : Hello world!
company.name               : My company

What the result is

name                             : My company
lastname                         : My Lastname
other_field                      : Hello world!

"customer.name" is identified as "name". "company.name" is also identified as "name".
The value is the latest found in document : "My Company".
--> value for "customer.name" is lost.

[PhunkyBob]

I switched to pypdf instead of PyPDF2.
The same problem occurs.

[MartinThoma]

Of course. It's the same software in the end. But in pypdf the development will continue and the issue will eventually be resolved.

[pubpub-zz]

@PhunkyBob,
I've implemented the handling of hierarchical fields, you will have to modify your code to add a true parameter to get_form_text_fields to get the fully_qualified_name

[PhunkyBob]

Thank you.
How can I test it?

[pubpub-zz]

Thank you.
How can I test it?

Yes you can load the files from the PR
It is not yet merged as waiting for adding test

[PhunkyBob]

The result with your hierachical_fields branch:

customer.name                    : My Name
customer.lastname                : My Lastname
other_field                      : Hello world!
company.name                     : My company

-> it's what I expected!

Thank you!
I hope this will be merged soon.

Code to reproduce:

from pypdf import PdfReader

if __name__ == "__main__":
    pdf_file_name = "fields_with_dots.pdf"
    pdf = PdfReader(pdf_file_name)
    fields = pdf.get_form_text_fields(full_qualified_name=True)
    for k, v in fields.items():
        print(f"{k:32} : {v}")

[ahmedshabib]

@PhunkyBob, I've implemented the handling of hierarchical fields, you will have to modify your code to add a true parameter to get_form_text_fields to get the fully_qualified_name

I checked your PR, and tried it locally, if I use get_fields , this gives me all of the keys now, but still has some stray keys.

[pubpub-zz]

I checked your PR, and tried it locally, if I use get_fields , this gives me all of the keys now, but still has some stray keys.

@ahmedshabib
Can you provide an pdf example withe "stray keys"?